Anxiety is among the most common concerns in children with autism spectrum disorder (ASD) who seek mental health services, causing suffering and family stress and exacerbating social disability. Cognitive- Behavioral Therapy (CBT) is a promising treatment for anxiety in children with high-functioning ASD, but the neural mechanisms supporting response to this treatment have not been studied. We will investigate the neural-systems-level targets of CBT for anxiety in children with ASD by evaluating brain indices of socio- emotional functioning with functional magnetic resonance imaging (fMRI) before and after treatment. CBT teaches emotion regulation skills, such as cognitive reappraisal, followed by behavioral exposure to anxiety- provoking situations. Because the goal of CBT is to replace automatic emotional reactivity with more controlled processing, we hypothesize that CBT aimed at the reduction of anxiety will enhance the neural circuitry that subserves the experience and regulation of emotions. This is a randomized controlled trial of CBT vs. Psychoeducation and Supportive Therapy (PST) in 100 school-age children with high-functioning ASD and moderate to severe anxiety. In addition, 50 matched typically-developing (TD) children will be scanned twice with a 16-week interval to enable interpretation of change in brain function in children with ASD after CBT vs. PST relative to TD children. CBT for anxiety will be provided using a structured manual that has been modified for children with ASD by increasing parental participation and addressing the role of core ASD symptoms in the experience and expression of anxiety. Subjects will be comprehensively characterized with regard to ASD diagnosis, IQ, adaptive behavior, and comorbid psychopathology. Primary measures of clinical outcomes will include the clinician-rated Pediatric Anxiety Rating Scale and the Clinical Global Impression-Improvement scale completed by an independent evaluator who will be blind to treatment assignment. Functional MRI will be collected as the subjects perform tasks involving emotion regulation, emotional face perception, biological motion perception, and resting state. We hypothesize that a positive response to CBT vs. PST, will be associated with increased activation of ventromedial and ventrolateral prefrontal cortices and increased amygdala-prefrontal functional connectivity during emotion regulation, along with reduced amygdala activation during emotional face perception. We will also examine the moderating effects of fMRI biomarkers of social functioning in ASD on response to CBT, and explore the utility of resting state fMRI in developing biomarkers of treatment response. This study will contribute to the understanding of the neuro circuitry of anxiety in ASD by 1) comparing children with ASD and anxiety to typically developing controls and 2) examining the association of reduction in anxiety after CBT vs. PST with changes in brain activity. Consistent with the IACC Strategic Plan for ASD Research, understanding of neural mechanisms will inform treatment of anxiety in ASD and improve the ability to select patients who are likely to benefit from CBT toward the goal of personalized care.